Modeling canopy photosynthesis from phloem sap δ13C and xylem sap flux: model development and application in two functionally distinct tree species
2 CBA, University of Lisbon
3 Agroecosystem/Functional Ecosystem Research, University of Bayreuth
P 3.11 in Ökosysteme: Funktion und Leistungen
In water limited ecosystems, where potential evapotranspiration exceeds precipitation, it is often assumed that plant invasions will not increase total ecosystem water use, since all available water is evaporated or transpired regardless of vegetation type. However, invasion by exotic species, with high water use rates, may potentially alter ecosystem water balance by reducing water available to native species, which may in turn impact carbon assimilation and productivity of co-occurring species. Here, we document the impact of invasion by an N2-fixing understory exotic woody species (Acacia longifolia) in a semi-arid Mediterranean dune Pine forest. To quantify the effects of this understory leguminous tree on the water use and carbon fixation rates of Pinus pinaster we compared an invaded and a non-invaded pine stand. We combined measurements of δ13C in various plant carbon pools (bulk leaf, leaf water soluble organic matter and phloem sap) with a continuous record of xylem sap flow (over a 10 month period encompassing a complete growing season).
In both species, we found significant post-photosynthetic fractionation as carbon was transported from the leaves towards the roots. These post-photosynthetic fractionation effects resulted in progressive depletion of δ13C in water-soluble organic matter pools along the plant axis from the canopy to the trunk (~6.5‰ depletion in A. longifolia and ~0.8‰ depletion in P. pinaster). Nevertheless, phloem sap δ13C in both terminal branches and the main stem correlated well with environmental parameters driving photosynthesis for both species, which allowed us to use phloem sap δ13C as an integrative tracer of changes in canopy carbon discrimination (Δ13C).
We developed a simple model combining xylem sap flow and phloem sap δ13C in order to estimate canopy-level carbon assimilation rates for both species over the entire growing season. Since δ13C signatures of phloem sap collected in the trunk integrate carbon fixed in the entire canopy they were used as an integrative proxy for changing canopy carbon isotope discrimination (Δ13C). Δ13C recorded in the phloem sap in May and August (spanning the wettest and driest time periods) was related to changes in vapor pressure deficit (VPD) and soil volumetric water content (VWC) using linear models. Models combining 4-day lagged VPD with same day soil VWC had the best predictive power. These models were then used to simulate Δ13C on a daily basis from changes in VPD and soil VWC throughout the entire study. Finally, canopy carbon assimilation rates (Acan) were estimated by applying Fick’s law to combine canopy stomatal conductance (GS), determined from xylem sap flow measurements using a modified Penman–Monteith equation, with the ratio of leaf internal to external CO2 concentration (ci/ca), determined from the simulated Δ13C using the Farquhar two-stage discrimination model for C3 species.
Both water use and carbon assimilation rates of P. pinaster were significantly reduced by A. longifolia invasion. A. longifolia contributed significantly to transpiration in the invaded forest (up to 42%) resulting in a slight increase (9%) in stand transpiration in the invaded relative to non-invaded forest. Furthermore, averaged over the entire study P. pinaster transpiration was 0.70 mm d-1 in the invaded and 0.93 mm d-1 in the non-invaded stand, a reduction of 25%. Furthermore, we show that on clear days across all seasons P. pinaster in the invaded forest had significantly lower canopy carbon assimilation rates compared to P. pinaster in the non-invaded stand.
Our study clearly shows that exotic plant invasions can have significant impacts on hydrological and carbon cycling even in water-limited semi-arid ecosystems through a repartitioning of water resources between the native and invasive species. Furthermore, the modeling technique described here has potential as a relatively simple, non-labor-intensive method for characterizing daily fluctuations in canopy carbon discrimination and carbon assimilation in other species and habitats.